In this paper, we explore the changes wrought by digitisation upon the domains of culture and knowledge. Half a century into the process of the digitisation of text, we argue that only now are we on the cusp of a series of paradigm shifts in the processes of writing, and concomitantly, our modes of cultural expression and our social processes of knowing. We describe the transition underway in the fundamental mechanics of rendering, the new navigational order which is associated with this transition, the demise of isolated written text that accompanies the rise of multimodality, the ubiquity of recording and documentation, a shift in the balance of representation agency, and its correlate in the emergence of a new dynamics of difference. The shape of these hugely significant changes is just beginning to become clear in the new, internet-mediated social media. The potential of the new textual regime is to transform our very means of production of meaning. However, when we come to examine the domain of formal knowledge production, historically pivoting on the peer reviewed journal and published monograph, there are as yet few signs of change. This paper points in a tentative way to potentials for knowledge-making which are as yet unrealised: new semantic markup processes which will improve knowledge discovery, data mining and machine translation; a new navigational order in which knowledge is not simply presented in a linear textual exegesis; the multimodal representation of knowledge in which knowledge evaluators and validators gain a broader, deeper and less mediated view of the knowledge they are assessing; navigable databanks in which reviewers and readers alike can make what they will of data and interactions recorded incidental to knowledge making; co-construction of knowledge through recursive dialogue between knowledge creators and knowledge users, to the extent of eliding that distinction; and a polylingual, polysemic knowledge world in which source natural language is arbitrary and narrowly specialised discourses and bodies of knowledge can be valued by their intellectual quality instead of the quantitative mass of their readership and citation.